Estimating the state of health (SOH) of a battery is one of the key features of a battery management system (BMS). A remaining challenge for most methods for SOH estimation is the change in the shape of the open circuit potential (OCP) curve of lithium-ion batteries, as this curve is often used to obtain reference values for the state of charge (SOC), which are needed for the SOH estimation. In this study, we present a method for SOH estimation based on current and voltage measurements during partial cell charging that does not require prior information on the SOC and takes into account changes in the shape of the OCP curve.

The change in the OCP curve of a lithium-ion cell can be described by changes in the OCP curves of the cell’s electrodes. A popular method for the diagnosis of cell aging is to align electrode OCP curves to aged full-cell OCP curves in order to obtain information on the aging modes that have occurred in the cell. In this study, we analyze whether this approach is also applicable to charging curves obtained at current rates that are typically used during the charging phase of battery electric vehicles. We also investigate the performance of the method if only partial charging curves are available, as it is often the case for the charging of battery electric vehicles. In order to show the performance of the method for different aging conditions, we apply it to charging data obtained for commercial cells aged under different conditions. The commercial cells used for this study contain silicon-graphite as anode active material and NMC-811 as cathode active material. As an additional aspect, we investigate if the SOH estimation can be improved by considering differences in the degradation rate of the anode blend components silicon and graphite.

The algorithm for SOH estimation presented in this study has several advantages in comparison to other methods: In addition to the estimate on the remaining capacity (SOH), an updated OCP curve and estimates for the aging modes that have occurred in the cell can be obtained. No prior information on the SOC is necessary and aging effects specific to silicon-graphite blend anodes are considered. Our results are relevant for both science and industry as silicon-graphite anodes are increasingly used in applications due to their high specific energy density.